The prevalence of obesity in the U.S. could rise sharply under a new definition of the condition released earlier this year by the Lancet Diabetes and Endocrinology Commission, according to research co-authored by Harvard-Mass General specialists.

Investigators from Harvard and Mass General Brigham found that when applying the new criteria, which expand upon the traditional use of body mass index (BMI) to include measures of body fat distribution, the prevalence of obesity increased from about 40 percent to about 70 percent among more than 300,000 people included in the study. The rise was more pronounced among older adults.

Additionally, the researchers found that the newly added individuals also had a higher risk of adverse health outcomes. The results are published in JAMA Network Open.

“We already thought we had an obesity epidemic, but this is astounding,” said co-first author Lindsay Fourman, a Mass General endocrinologist and an assistant professor at Harvard Medical School. “With potentially 70 percent of the adult population now considered to have excess fat, we need to better understand what treatment approaches to prioritize.”

Traditionally, obesity has been defined by BMI, which estimates body fat based on a person’s weight and height. But other anthropomorphic measures — such as waist circumference, waist-to-height ratio, or waist-to-hip ratio — may further account for fat distribution and aid in differentiation between muscle and fat mass.

Under the new framework, a person is classified as having obesity if they have a high BMI plus at least one elevated anthropometric measure (a condition the authors term “BMI-plus-anthropometric obesity”), or if they have a normal BMI and at least two elevated anthropometric measures (a condition termed “anthropometric-only obesity”).

The new definition also distinguishes between preclinical and clinical obesity, with clinical obesity defined as the presence of obesity-related physical impairment or organ dysfunction. At least 76 organizations have endorsed the new guidelines, including the American Heart Association and the Obesity Society.

The study analyzed participants in the National Institutes of Health All of Us Research Program’s cohort of more than 300,000 Americans. Obesity prevalence was 68.6 percent with the new definition, versus 42.9 percent under the traditional BMI-based definition. This increase was entirely driven by inclusion of individuals with anthropometric-only obesity. Obesity rates varied by sex, race, and especially by age — affecting nearly 80 percent of adults over 70.

Importantly, the study found that those with anthropometric-only obesity — who would not have been classified as having obesity by the traditional definition — had a higher risk of diabetes, cardiovascular disease, and mortality than people without obesity. About half of all individuals who met the new obesity criteria had clinical obesity, and this proportion was only slightly lower in the anthropometric-only obesity group compared with the BMI-plus-anthropometric obesity group.

“We have always recognized the limitations of BMI as a single marker for obesity because it doesn’t take into account body fat distribution,” said senior author Steven Grinspoon of Mass General and Harvard Medical School. “Seeing an increased risk of cardiovascular disease and diabetes in this new group of people with obesity, who were not considered to have obesity before, brings up interesting questions about obesity medications and other therapeutics.”

The researchers emphasized that further studies are needed to better understand the causes of and optimal treatments for anthropometric-only obesity. The team previously developed a therapeutic that reduces waist circumference, and plans to explore the utility of different treatment strategies in this newly defined population.

“Identifying excess body fat is very important as we’re finding that even people with a normal BMI but with abdominal fat accumulation are at increased health risk,” Fourman said. “Body composition matters — it’s not just pounds on a scale.”

The research described in this article received funding from the National Institutes of Health.